1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
|
doctests = """
Tests for the tokenize module.
>>> import glob, random, sys
The tests can be really simple. Given a small fragment of source
code, print out a table with thokens. The ENDMARK is omitted for
brevity.
>>> dump_tokens("1 + 1")
NUMBER '1' (1, 0) (1, 1)
OP '+' (1, 2) (1, 3)
NUMBER '1' (1, 4) (1, 5)
>>> dump_tokens("if False:\\n"
... " # NL\\n"
... " True = False # NEWLINE\\n")
NAME 'if' (1, 0) (1, 2)
NAME 'False' (1, 3) (1, 8)
OP ':' (1, 8) (1, 9)
NEWLINE '\\n' (1, 9) (1, 10)
COMMENT '# NL' (2, 4) (2, 8)
NL '\\n' (2, 8) (2, 9)
INDENT ' ' (3, 0) (3, 4)
NAME 'True' (3, 4) (3, 8)
OP '=' (3, 9) (3, 10)
NAME 'False' (3, 11) (3, 16)
COMMENT '# NEWLINE' (3, 17) (3, 26)
NEWLINE '\\n' (3, 26) (3, 27)
DEDENT '' (4, 0) (4, 0)
>>> indent_error_file = \"""
... def k(x):
... x += 2
... x += 5
... \"""
>>> for tok in generate_tokens(StringIO(indent_error_file).readline): pass
Traceback (most recent call last):
...
IndentationError: unindent does not match any outer indentation level
Test roundtrip for `untokenize`. `f` is an open file or a string. The source
code in f is tokenized, converted back to source code via tokenize.untokenize(),
and tokenized again from the latter. The test fails if the second tokenization
doesn't match the first.
>>> def roundtrip(f):
... if isinstance(f, str): f = StringIO(f)
... token_list = list(generate_tokens(f.readline))
... f.close()
... tokens1 = [tok[:2] for tok in token_list]
... new_text = untokenize(tokens1)
... readline = iter(new_text.splitlines(1)).__next__
... tokens2 = [tok[:2] for tok in generate_tokens(readline)]
... return tokens1 == tokens2
...
There are some standard formattig practises that are easy to get right.
>>> roundtrip("if x == 1:\\n"
... " print(x)\\n")
True
>>> roundtrip("# This is a comment\\n# This also")
True
Some people use different formatting conventions, which makes
untokenize a little trickier. Note that this test involves trailing
whitespace after the colon. Note that we use hex escapes to make the
two trailing blanks apperant in the expected output.
>>> roundtrip("if x == 1 : \\n"
... " print(x)\\n")
True
>>> f = test_support.findfile("tokenize_tests.txt")
>>> roundtrip(open(f))
True
>>> roundtrip("if x == 1:\\n"
... " # A comment by itself.\\n"
... " print(x) # Comment here, too.\\n"
... " # Another comment.\\n"
... "after_if = True\\n")
True
>>> roundtrip("if (x # The comments need to go in the right place\\n"
... " == 1):\\n"
... " print('x==1')\\n")
True
>>> roundtrip("class Test: # A comment here\\n"
... " # A comment with weird indent\\n"
... " after_com = 5\\n"
... " def x(m): return m*5 # a one liner\\n"
... " def y(m): # A whitespace after the colon\\n"
... " return y*4 # 3-space indent\\n")
True
Some error-handling code
>>> roundtrip("try: import somemodule\\n"
... "except ImportError: # comment\\n"
... " print 'Can not import' # comment2\\n"
... "else: print 'Loaded'\\n")
True
Balancing contunuation
>>> roundtrip("a = (3,4, \\n"
... "5,6)\\n"
... "y = [3, 4,\\n"
... "5]\\n"
... "z = {'a': 5,\\n"
... "'b':15, 'c':True}\\n"
... "x = len(y) + 5 - a[\\n"
... "3] - a[2]\\n"
... "+ len(z) - z[\\n"
... "'b']\\n")
True
Ordinary integers and binary operators
>>> dump_tokens("0xff <= 255")
NUMBER '0xff' (1, 0) (1, 4)
OP '<=' (1, 5) (1, 7)
NUMBER '255' (1, 8) (1, 11)
>>> dump_tokens("01234567 > ~0x15")
NUMBER '01234567' (1, 0) (1, 8)
OP '>' (1, 9) (1, 10)
OP '~' (1, 11) (1, 12)
NUMBER '0x15' (1, 12) (1, 16)
>>> dump_tokens("2134568 != 01231515")
NUMBER '2134568' (1, 0) (1, 7)
OP '!=' (1, 8) (1, 10)
NUMBER '01231515' (1, 11) (1, 19)
>>> dump_tokens("(-124561-1) & 0200000000")
OP '(' (1, 0) (1, 1)
OP '-' (1, 1) (1, 2)
NUMBER '124561' (1, 2) (1, 8)
OP '-' (1, 8) (1, 9)
NUMBER '1' (1, 9) (1, 10)
OP ')' (1, 10) (1, 11)
OP '&' (1, 12) (1, 13)
NUMBER '0200000000' (1, 14) (1, 24)
>>> dump_tokens("0xdeadbeef != -1")
NUMBER '0xdeadbeef' (1, 0) (1, 10)
OP '!=' (1, 11) (1, 13)
OP '-' (1, 14) (1, 15)
NUMBER '1' (1, 15) (1, 16)
>>> dump_tokens("0xdeadc0de & 012345")
NUMBER '0xdeadc0de' (1, 0) (1, 10)
OP '&' (1, 11) (1, 12)
NUMBER '012345' (1, 13) (1, 19)
>>> dump_tokens("0xFF & 0x15 | 1234")
NUMBER '0xFF' (1, 0) (1, 4)
OP '&' (1, 5) (1, 6)
NUMBER '0x15' (1, 7) (1, 11)
OP '|' (1, 12) (1, 13)
NUMBER '1234' (1, 14) (1, 18)
Long integers
>>> dump_tokens("x = 0L")
NAME 'x' (1, 0) (1, 1)
OP '=' (1, 2) (1, 3)
NUMBER '0L' (1, 4) (1, 6)
>>> dump_tokens("x = 0xfffffffffff")
NAME 'x' (1, 0) (1, 1)
OP '=' (1, 2) (1, 3)
NUMBER '0xffffffffff (1, 4) (1, 17)
>>> dump_tokens("x = 123141242151251616110l")
NAME 'x' (1, 0) (1, 1)
OP '=' (1, 2) (1, 3)
NUMBER '123141242151 (1, 4) (1, 26)
>>> dump_tokens("x = -15921590215012591L")
NAME 'x' (1, 0) (1, 1)
OP '=' (1, 2) (1, 3)
OP '-' (1, 4) (1, 5)
NUMBER '159215902150 (1, 5) (1, 23)
Floating point numbers
>>> dump_tokens("x = 3.14159")
NAME 'x' (1, 0) (1, 1)
OP '=' (1, 2) (1, 3)
NUMBER '3.14159' (1, 4) (1, 11)
>>> dump_tokens("x = 314159.")
NAME 'x' (1, 0) (1, 1)
OP '=' (1, 2) (1, 3)
NUMBER '314159.' (1, 4) (1, 11)
>>> dump_tokens("x = .314159")
NAME 'x' (1, 0) (1, 1)
OP '=' (1, 2) (1, 3)
NUMBER '.314159' (1, 4) (1, 11)
>>> dump_tokens("x = 3e14159")
NAME 'x' (1, 0) (1, 1)
OP '=' (1, 2) (1, 3)
NUMBER '3e14159' (1, 4) (1, 11)
>>> dump_tokens("x = 3E123")
NAME 'x' (1, 0) (1, 1)
OP '=' (1, 2) (1, 3)
NUMBER '3E123' (1, 4) (1, 9)
>>> dump_tokens("x+y = 3e-1230")
NAME 'x' (1, 0) (1, 1)
OP '+' (1, 1) (1, 2)
NAME 'y' (1, 2) (1, 3)
OP '=' (1, 4) (1, 5)
NUMBER '3e-1230' (1, 6) (1, 13)
>>> dump_tokens("x = 3.14e159")
NAME 'x' (1, 0) (1, 1)
OP '=' (1, 2) (1, 3)
NUMBER '3.14e159' (1, 4) (1, 12)
String literals
>>> dump_tokens("x = ''; y = \\\"\\\"")
NAME 'x' (1, 0) (1, 1)
OP '=' (1, 2) (1, 3)
STRING "''" (1, 4) (1, 6)
OP ';' (1, 6) (1, 7)
NAME 'y' (1, 8) (1, 9)
OP '=' (1, 10) (1, 11)
STRING '""' (1, 12) (1, 14)
>>> dump_tokens("x = '\\\"'; y = \\\"'\\\"")
NAME 'x' (1, 0) (1, 1)
OP '=' (1, 2) (1, 3)
STRING '\\'"\\'' (1, 4) (1, 7)
OP ';' (1, 7) (1, 8)
NAME 'y' (1, 9) (1, 10)
OP '=' (1, 11) (1, 12)
STRING '"\\'"' (1, 13) (1, 16)
>>> dump_tokens("x = \\\"doesn't \\\"shrink\\\", does it\\\"")
NAME 'x' (1, 0) (1, 1)
OP '=' (1, 2) (1, 3)
STRING '"doesn\\'t "' (1, 4) (1, 14)
NAME 'shrink' (1, 14) (1, 20)
STRING '", does it"' (1, 20) (1, 31)
>>> dump_tokens("x = u'abc' + U'ABC'")
NAME 'x' (1, 0) (1, 1)
OP '=' (1, 2) (1, 3)
STRING "u'abc'" (1, 4) (1, 10)
OP '+' (1, 11) (1, 12)
STRING "U'ABC'" (1, 13) (1, 19)
>>> dump_tokens('y = "ABC" + "ABC"')
NAME 'y' (1, 0) (1, 1)
OP '=' (1, 2) (1, 3)
STRING '"ABC"' (1, 4) (1, 10)
OP '+' (1, 11) (1, 12)
STRING '"ABC"' (1, 13) (1, 19)
>>> dump_tokens("x = r'abc' + r'ABC' + R'ABC' + R'ABC'")
NAME 'x' (1, 0) (1, 1)
OP '=' (1, 2) (1, 3)
STRING "r'abc'" (1, 4) (1, 11)
OP '+' (1, 12) (1, 13)
STRING "r'ABC'" (1, 14) (1, 21)
OP '+' (1, 22) (1, 23)
STRING "R'ABC'" (1, 24) (1, 31)
OP '+' (1, 32) (1, 33)
STRING "R'ABC'" (1, 34) (1, 41)
>>> dump_tokens('y = r"abc" + r"ABC" + R"ABC" + R"ABC"')
NAME 'y' (1, 0) (1, 1)
OP '=' (1, 2) (1, 3)
STRING 'r"abc"' (1, 4) (1, 11)
OP '+' (1, 12) (1, 13)
STRING 'r"ABC"' (1, 14) (1, 21)
OP '+' (1, 22) (1, 23)
STRING 'R"ABC"' (1, 24) (1, 31)
OP '+' (1, 32) (1, 33)
STRING 'R"ABC"' (1, 34) (1, 41)
Operators
>>> dump_tokens("def d22(a, b, c=2, d=2, *k): pass")
NAME 'def' (1, 0) (1, 3)
NAME 'd22' (1, 4) (1, 7)
OP '(' (1, 7) (1, 8)
NAME 'a' (1, 8) (1, 9)
OP ',' (1, 9) (1, 10)
NAME 'b' (1, 11) (1, 12)
OP ',' (1, 12) (1, 13)
NAME 'c' (1, 14) (1, 15)
OP '=' (1, 15) (1, 16)
NUMBER '2' (1, 16) (1, 17)
OP ',' (1, 17) (1, 18)
NAME 'd' (1, 19) (1, 20)
OP '=' (1, 20) (1, 21)
NUMBER '2' (1, 21) (1, 22)
OP ',' (1, 22) (1, 23)
OP '*' (1, 24) (1, 25)
NAME 'k' (1, 25) (1, 26)
OP ')' (1, 26) (1, 27)
OP ':' (1, 27) (1, 28)
NAME 'pass' (1, 29) (1, 33)
>>> dump_tokens("def d01v_(a=1, *k, **w): pass")
NAME 'def' (1, 0) (1, 3)
NAME 'd01v_' (1, 4) (1, 9)
OP '(' (1, 9) (1, 10)
NAME 'a' (1, 10) (1, 11)
OP '=' (1, 11) (1, 12)
NUMBER '1' (1, 12) (1, 13)
OP ',' (1, 13) (1, 14)
OP '*' (1, 15) (1, 16)
NAME 'k' (1, 16) (1, 17)
OP ',' (1, 17) (1, 18)
OP '**' (1, 19) (1, 21)
NAME 'w' (1, 21) (1, 22)
OP ')' (1, 22) (1, 23)
OP ':' (1, 23) (1, 24)
NAME 'pass' (1, 25) (1, 29)
Comparison
>>> dump_tokens("if 1 < 1 > 1 == 1 >= 5 <= 0x15 <= 0x12 != " +
... "1 and 5 in 1 not in 1 is 1 or 5 is not 1: pass")
NAME 'if' (1, 0) (1, 2)
NUMBER '1' (1, 3) (1, 4)
OP '<' (1, 5) (1, 6)
NUMBER '1' (1, 7) (1, 8)
OP '>' (1, 9) (1, 10)
NUMBER '1' (1, 11) (1, 12)
OP '==' (1, 13) (1, 15)
NUMBER '1' (1, 16) (1, 17)
OP '>=' (1, 18) (1, 20)
NUMBER '5' (1, 21) (1, 22)
OP '<=' (1, 23) (1, 25)
NUMBER '0x15' (1, 26) (1, 30)
OP '<=' (1, 31) (1, 33)
NUMBER '0x12' (1, 34) (1, 38)
OP '!=' (1, 39) (1, 41)
NUMBER '1' (1, 42) (1, 43)
NAME 'and' (1, 44) (1, 47)
NUMBER '5' (1, 48) (1, 49)
NAME 'in' (1, 50) (1, 52)
NUMBER '1' (1, 53) (1, 54)
NAME 'not' (1, 55) (1, 58)
NAME 'in' (1, 59) (1, 61)
NUMBER '1' (1, 62) (1, 63)
NAME 'is' (1, 64) (1, 66)
NUMBER '1' (1, 67) (1, 68)
NAME 'or' (1, 69) (1, 71)
NUMBER '5' (1, 72) (1, 73)
NAME 'is' (1, 74) (1, 76)
NAME 'not' (1, 77) (1, 80)
NUMBER '1' (1, 81) (1, 82)
OP ':' (1, 82) (1, 83)
NAME 'pass' (1, 84) (1, 88)
Shift
>>> dump_tokens("x = 1 << 1 >> 5")
NAME 'x' (1, 0) (1, 1)
OP '=' (1, 2) (1, 3)
NUMBER '1' (1, 4) (1, 5)
OP '<<' (1, 6) (1, 8)
NUMBER '1' (1, 9) (1, 10)
OP '>>' (1, 11) (1, 13)
NUMBER '5' (1, 14) (1, 15)
Additive
>>> dump_tokens("x = 1 - y + 15 - 01 + 0x124 + z + a[5]")
NAME 'x' (1, 0) (1, 1)
OP '=' (1, 2) (1, 3)
NUMBER '1' (1, 4) (1, 5)
OP '-' (1, 6) (1, 7)
NAME 'y' (1, 8) (1, 9)
OP '+' (1, 10) (1, 11)
NUMBER '15' (1, 12) (1, 14)
OP '-' (1, 15) (1, 16)
NUMBER '01' (1, 17) (1, 19)
OP '+' (1, 20) (1, 21)
NUMBER '0x124' (1, 22) (1, 27)
OP '+' (1, 28) (1, 29)
NAME 'z' (1, 30) (1, 31)
OP '+' (1, 32) (1, 33)
NAME 'a' (1, 34) (1, 35)
OP '[' (1, 35) (1, 36)
NUMBER '5' (1, 36) (1, 37)
OP ']' (1, 37) (1, 38)
Multiplicative
>>> dump_tokens("x = 1//1*1/5*12%0x12")
NAME 'x' (1, 0) (1, 1)
OP '=' (1, 2) (1, 3)
NUMBER '1' (1, 4) (1, 5)
OP '//' (1, 5) (1, 7)
NUMBER '1' (1, 7) (1, 8)
OP '*' (1, 8) (1, 9)
NUMBER '1' (1, 9) (1, 10)
OP '/' (1, 10) (1, 11)
NUMBER '5' (1, 11) (1, 12)
OP '*' (1, 12) (1, 13)
NUMBER '12' (1, 13) (1, 15)
OP '%' (1, 15) (1, 16)
NUMBER '0x12' (1, 16) (1, 20)
Unary
>>> dump_tokens("~1 ^ 1 & 1 |1 ^ -1")
OP '~' (1, 0) (1, 1)
NUMBER '1' (1, 1) (1, 2)
OP '^' (1, 3) (1, 4)
NUMBER '1' (1, 5) (1, 6)
OP '&' (1, 7) (1, 8)
NUMBER '1' (1, 9) (1, 10)
OP '|' (1, 11) (1, 12)
NUMBER '1' (1, 12) (1, 13)
OP '^' (1, 14) (1, 15)
OP '-' (1, 16) (1, 17)
NUMBER '1' (1, 17) (1, 18)
>>> dump_tokens("-1*1/1+1*1//1 - ---1**1")
OP '-' (1, 0) (1, 1)
NUMBER '1' (1, 1) (1, 2)
OP '*' (1, 2) (1, 3)
NUMBER '1' (1, 3) (1, 4)
OP '/' (1, 4) (1, 5)
NUMBER '1' (1, 5) (1, 6)
OP '+' (1, 6) (1, 7)
NUMBER '1' (1, 7) (1, 8)
OP '*' (1, 8) (1, 9)
NUMBER '1' (1, 9) (1, 10)
OP '//' (1, 10) (1, 12)
NUMBER '1' (1, 12) (1, 13)
OP '-' (1, 14) (1, 15)
OP '-' (1, 16) (1, 17)
OP '-' (1, 17) (1, 18)
OP '-' (1, 18) (1, 19)
NUMBER '1' (1, 19) (1, 20)
OP '**' (1, 20) (1, 22)
NUMBER '1' (1, 22) (1, 23)
Selector
>>> dump_tokens("import sys, time\\nx = sys.modules['time'].time()")
NAME 'import' (1, 0) (1, 6)
NAME 'sys' (1, 7) (1, 10)
OP ',' (1, 10) (1, 11)
NAME 'time' (1, 12) (1, 16)
NEWLINE '\\n' (1, 16) (1, 17)
NAME 'x' (2, 0) (2, 1)
OP '=' (2, 2) (2, 3)
NAME 'sys' (2, 4) (2, 7)
OP '.' (2, 7) (2, 8)
NAME 'modules' (2, 8) (2, 15)
OP '[' (2, 15) (2, 16)
STRING "'time'" (2, 16) (2, 22)
OP ']' (2, 22) (2, 23)
OP '.' (2, 23) (2, 24)
NAME 'time' (2, 24) (2, 28)
OP '(' (2, 28) (2, 29)
OP ')' (2, 29) (2, 30)
Methods
>>> dump_tokens("@staticmethod\\ndef foo(x,y): pass")
OP '@' (1, 0) (1, 1)
NAME 'staticmethod (1, 1) (1, 13)
NEWLINE '\\n' (1, 13) (1, 14)
NAME 'def' (2, 0) (2, 3)
NAME 'foo' (2, 4) (2, 7)
OP '(' (2, 7) (2, 8)
NAME 'x' (2, 8) (2, 9)
OP ',' (2, 9) (2, 10)
NAME 'y' (2, 10) (2, 11)
OP ')' (2, 11) (2, 12)
OP ':' (2, 12) (2, 13)
NAME 'pass' (2, 14) (2, 18)
Backslash means line continuation, except for comments
>>> roundtrip("x=1+\\\\n"
... "1\\n"
... "# This is a comment\\\\n"
... "# This also\\n")
True
>>> roundtrip("# Comment \\\\nx = 0")
True
>>>
>>> tempdir = os.path.dirname(f) or os.curdir
>>> testfiles = glob.glob(os.path.join(tempdir, "test*.py"))
>>> if not test_support.is_resource_enabled("compiler"):
... testfiles = random.sample(testfiles, 10)
...
>>> for testfile in testfiles:
... if not roundtrip(open(testfile)): break
... else: True
True
"""
from test import test_support
from tokenize import (tokenize, untokenize, generate_tokens, NUMBER, NAME, OP,
STRING, ENDMARKER, tok_name)
from io import StringIO
import os
def dump_tokens(s):
"""Print out the tokens in s in a table format.
The ENDMARKER is omitted.
"""
f = StringIO(s)
for type, token, start, end, line in generate_tokens(f.readline):
if type == ENDMARKER:
break
type = tok_name[type]
print("%(type)-10.10s %(token)-13.13r %(start)s %(end)s" % locals())
def roundtrip(s):
f = StringIO(s)
source = untokenize(generate_tokens(f.readline))
print(source, end="")
# This is an example from the docs, set up as a doctest.
def decistmt(s):
"""Substitute Decimals for floats in a string of statements.
>>> from decimal import Decimal
>>> s = 'print(+21.3e-5*-.1234/81.7)'
>>> decistmt(s)
"print (+Decimal ('21.3e-5')*-Decimal ('.1234')/Decimal ('81.7'))"
The format of the exponent is inherited from the platform C library.
Known cases are "e-007" (Windows) and "e-07" (not Windows). Since
we're only showing 12 digits, and the 13th isn't close to 5, the
rest of the output should be platform-independent.
>>> exec(s) #doctest: +ELLIPSIS
-3.21716034272e-0...7
Output from calculations with Decimal should be identical across all
platforms.
>>> exec(decistmt(s))
-3.217160342717258261933904529E-7
"""
result = []
g = generate_tokens(StringIO(s).readline) # tokenize the string
for toknum, tokval, _, _, _ in g:
if toknum == NUMBER and '.' in tokval: # replace NUMBER tokens
result.extend([
(NAME, 'Decimal'),
(OP, '('),
(STRING, repr(tokval)),
(OP, ')')
])
else:
result.append((toknum, tokval))
return untokenize(result)
__test__ = {"doctests" : doctests, 'decistmt': decistmt}
def test_main():
from test import test_tokenize
test_support.run_doctest(test_tokenize, True)
if __name__ == "__main__":
test_main()
|